Port more test cases to the DSL with SORT, STAT and $CORES

[jviotti]

Signed-off-by: Juan Cruz Viotti jv@jviotti.com

[augmentcode]

🤖 Augment PR Summary

Summary: This PR ports additional shell-based CLI tests to the Python clitest DSL.

Changes:

• Adds DSL primitives for sorting file contents and recording nanosecond modification times.
• Introduces a reserved $CORES variable for asserting default parallelism output.
• Migrates formatter mtime, cURL-loading, and multi-job test scenarios from shell scripts.
• Registers the migrated tests through the platform-neutral CMake test helper.
• Updates the missing-cURL CI test to use the DSL under its existing Linux gate.

Technical Notes:

• Parallel-output tests normalize paths and sort output where ordering is intentionally nondeterministic.
• Error-output assertions retain the DSL's stdout/stderr prefixes for exact comparison.

_{🤖 Was this summary useful? React with 👍 or 👎}

[augmentcode]

Review completed. 2 suggestions posted.

Comment augment review to trigger a new review at any time.

[augmentcode]

test/format/pass_single_intact_mtime.clitest:3: This assumes nanosecond mtime resolution, but this test now runs on every platform and temporary directories can be backed by filesystems with coarser resolution. A formatter that rewrites the identical file within the same timestamp tick would still pass, so the migrated test no longer reliably enforces the no-rewrite guarantee.

Severity: low

_{🤖 Was this useful? React with 👍 or 👎, or 🚀 if it prevented an incident/outage.}

[cubic-dev-ai]

1 issue found across 18 files

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="test/format/pass_single_intact_mtime.clitest">

<violation number="1" location="test/format/pass_single_intact_mtime.clitest:16">
P3: This test now runs on every platform (previously UNIX-only), but it relies on nanosecond mtime resolution via STAT MTIME without any delay between writes. On filesystems or platforms with coarser mtime granularity, a formatter that rewrites the identical file within the same timestamp tick would still pass this check, so the migrated test may no longer reliably enforce the no-rewrite guarantee that the original shell test (with its 1-second sleep) guaranteed.</violation>
</file>

<sub>Reply with feedback, questions, or to request a fix.

Re-trigger cubic</sub>

[cubic-dev-ai]

P3: This test now runs on every platform (previously UNIX-only), but it relies on nanosecond mtime resolution via STAT MTIME without any delay between writes. On filesystems or platforms with coarser mtime granularity, a formatter that rewrites the identical file within the same timestamp tick would still pass this check, so the migrated test may no longer reliably enforce the no-rewrite guarantee that the original shell test (with its 1-second sleep) guaranteed.

Prompt for AI agents

Check if this issue is valid — if so, understand the root cause and fix it. At test/format/pass_single_intact_mtime.clitest, line 16:

<comment>This test now runs on every platform (previously UNIX-only), but it relies on nanosecond mtime resolution via STAT MTIME without any delay between writes. On filesystems or platforms with coarser mtime granularity, a formatter that rewrites the identical file within the same timestamp tick would still pass this check, so the migrated test may no longer reliably enforce the no-rewrite guarantee that the original shell test (with its 1-second sleep) guaranteed.</comment>

<file context>
@@ -0,0 +1,30 @@
+
+COPY schema.json TO original.json
+
+STAT MTIME schema.json INTO mtime_before.txt
+
+RUN fmt schema.json STDIN /dev/null IN . INTO result_0.txt EXPECTING 0
</file context>